c32ac5b0d5
fix references for AdamW
robintibor
2019-01-07 15:48:44 +01:00
6c3ead2449
fixed bug in split train test
Robin Tibor Schirrmeister
2018-10-29 20:57:35 +01:00
cab5840ecf
added padding option to avgpool2d with conv, fixed expression str in some situations
Robin Tibor Schirrmeister
2018-10-26 17:35:43 +02:00
96ee7839b4
changed local data folder
Robin Tibor Schirrmeister
2018-10-26 17:35:10 +02:00
651ef3d7a4
fix using cuda with model.predict
Robin Tibor Schirrmeister
2018-09-19 12:33:20 +02:00
2760799fc0
fix reference
Robin Tibor Schirrmeister
2018-09-19 12:32:48 +02:00
2d9059cec5
new docs
gh-pages
Robin Tibor Schirrmeister
2018-09-17 17:41:20 +02:00
895b03703e
increment version
Robin Tibor Schirrmeister
2018-09-17 17:38:04 +02:00
5ea159b190
adapt base model for already dense model
Robin Tibor Schirrmeister
2018-09-17 17:23:13 +02:00
81678c4ab5
split hybrid into pytorch module and our model
Robin Tibor Schirrmeister
2018-09-17 17:23:04 +02:00
21890f9e19
changed logic to always creting random state in function body; not in function signature
Robin Tibor Schirrmeister
2018-09-17 17:20:21 +02:00
5f7e7fa3e4
mentioning adamw
Robin Tibor Schirrmeister
2018-09-17 17:19:35 +02:00
d1bf8a5ca1
upgrade pytorch 0.4
Robin Tibor Schirrmeister
2018-09-17 17:19:26 +02:00
82e6760a54
splitting into train/valid/test by folds
Robin Tibor Schirrmeister
2018-09-17 17:18:47 +02:00
fb080fc9b9
removed seed from experiment class
Robin Tibor Schirrmeister
2018-09-10 14:24:18 +02:00
13b791d965
updated references
Robin Tibor Schirrmeister
2018-09-10 14:23:56 +02:00
1e3fcca4f8
updated references
Robin Tibor Schirrmeister
2018-09-03 17:07:57 +02:00
d01feb6fa5
merge
Robin Tibor Schirrmeister
2018-09-03 16:14:06 +02:00
b69a930799
put hyperparameters into one place in code
Robin Tibor Schirrmeister
2018-08-27 14:22:24 +02:00
b0f186e3e0
some change to notebookdocs
Robin Tibor Schirrmeister
2018-08-27 14:21:22 +02:00
da413aae36
fixed shuffle=False msitake in cropped training
Robin Tibor Schirrmeister
2018-08-27 12:47:16 +02:00
6a1c9bf588
added missing paranthesis in if statement. fixed logging of resetting to best eopch after second run
Lukas gemein
2018-07-18 14:42:49 +02:00
069e8e36d3
only resetting to epoch with best valid_misclass if valid_loss did not deceed best 1st run train_loss
Lukas gemein
2018-07-17 18:06:04 +02:00
b7f3ee055f
enabled remembering best epoch for second run after early stopping. if in the second run no valid_loss was found that deceeds the best train_loss, reset the model to the epoch with the lowest valid_misclass
Lukas gemein
2018-07-17 17:54:19 +02:00
f84b444547
fixed a bug causing the rng of ClassBalancedBatchSizeIterator to be reset to a state different from the initial random state. added seeding parameter to BalancedBatchSizeIterator and ClassBalancedBatchSizeIterator. added seeding parameter to experiment class to allow control of torch and cuda random state.
Lukas gemein
2018-07-13 11:03:03 +02:00
d81caee2c7
next version
Robin Tibor Schirrmeister
2018-06-12 17:20:11 +02:00
c4d0dea093
next version
Robin Tibor Schirrmeister
2018-06-12 17:17:28 +02:00
fe260e2627
filtfilt option
Robin Tibor Schirrmeister
2018-06-12 17:16:38 +02:00
0bfc5f2f3d
factored out convert angle to cart function
Robin Tibor Schirrmeister
2018-06-12 17:16:09 +02:00
c7354c8d03
Merge branch 'master' of github.com:robintibor/braindecode
Robin Tibor Schirrmeister
2018-06-12 17:14:08 +02:00
c3b987bbf3
rename fixed weights to pool_weights in AvgPool2dWithConv
Robin Tibor Schirrmeister
2018-06-12 17:14:04 +02:00
ff9ae54b75
update bbci to not use dat field anymore, just nfo
Robin Tibor Schirrmeister
2018-06-12 17:13:07 +02:00
6e68691952
ensure makers are ints
robintibor
2018-03-22 12:21:49 +01:00
5fd38fb9b0
Sinus fits
Kay Gregor Hartmann
2018-02-07 22:19:24 +01:00
f33e2e3738
Most activating input windows and reverse correlation
Kay Gregor Hartmann
2018-02-07 21:39:45 +01:00
c141351cf0
Functions for determining and retrieving maximising input windows
Kay Gregor Hartmann
2018-02-07 17:00:10 +01:00
bd928df0d0
changes
Kay Gregor Hartmann
2018-02-07 16:28:16 +01:00
21870f5d55
Faster feature map correlation
Kay Gregor Hartmann
2018-01-26 08:41:00 +01:00
7a7be4326b
Fixes
Kay Gregor Hartmann
2018-01-26 08:27:09 +01:00
83acd8c3e9
Doc fix
Kay Gregor Hartmann
2018-01-26 07:35:11 +01:00
51fde3f31c
Use new spectral_perturbation_correlation function for old compute_amplitude_prediction_correlations
Kay Gregor Hartmann
2018-01-26 07:30:54 +01:00
8049193651
Perturbation functions Kay
Kay Gregor Hartmann
2018-01-25 14:48:13 +01:00
56c21702ad
fixed typo and 0.3.1
Robin Tibor Schirrmeister
2017-11-27 13:18:38 +01:00
86971b8ff8
do early stop option, seed for iterator, named functions for networks for pickle possibility
Robin Tibor Schirrmeister
2017-11-27 13:12:04 +01:00
d5d08a2ea3
updated installation instructions
Robin Tibor Schirrmeister
2017-11-03 18:01:32 +01:00
451cbccebe
notebooks with output?
Robin Tibor Schirrmeister
2017-11-03 17:59:50 +01:00
86b1df8490
mne separate install no longer necessary
Robin Tibor Schirrmeister
2017-11-03 17:47:03 +01:00
3f809ddbeb
monitors for binary case, experiment works with lists of outputs also
Robin Tibor Schirrmeister
2017-11-03 17:19:32 +01:00
9a87f29f17
segmentation changed
Robin Tibor Schirrmeister
2017-10-18 15:35:15 +02:00
5b5baf6c54
Merge branch 'master' of github.com:robintibor/braindecode
Robin Tibor Schirrmeister
2017-10-18 15:33:24 +02:00
750e060f3d
now all trial segmentation happens by creating continuous labels first
Robin Tibor Schirrmeister
2017-10-18 15:33:12 +02:00
1a77e49a5b
add breaks back to +1/-1 after before trial logic
Robin Tibor Schirrmeister
2017-10-18 11:51:07 +02:00
f424e66674
more tiral segment logic, before removing old logic
Robin Tibor Schirrmeister
2017-10-18 11:24:17 +02:00